datadog-ruby_core_source 3.4.1 → 3.5.0

data/lib/datadog/ruby_core_source/ruby-4.0.0-preview2/prism/util/pm_newline_list.h

@@ -0,0 +1,113 @@
+/**
+ * @file pm_newline_list.h
+ *
+ * A list of byte offsets of newlines in a string.
+ *
+ * When compiling the syntax tree, it's necessary to know the line and column
+ * of many nodes. This is necessary to support things like error messages,
+ * tracepoints, etc.
+ *
+ * It's possible that we could store the start line, start column, end line, and
+ * end column on every node in addition to the offsets that we already store,
+ * but that would be quite a lot of memory overhead.
+ */
+#ifndef PRISM_NEWLINE_LIST_H
+#define PRISM_NEWLINE_LIST_H
+
+#include "prism/defines.h"
+
+#include <assert.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdlib.h>
+
+/**
+ * A list of offsets of newlines in a string. The offsets are assumed to be
+ * sorted/inserted in ascending order.
+ */
+typedef struct {
+    /** A pointer to the start of the source string. */
+    const uint8_t *start;
+
+    /** The number of offsets in the list. */
+    size_t size;
+
+    /** The capacity of the list that has been allocated. */
+    size_t capacity;
+
+    /** The list of offsets. */
+    size_t *offsets;
+} pm_newline_list_t;
+
+/**
+ * A line and column in a string.
+ */
+typedef struct {
+    /** The line number. */
+    int32_t line;
+
+    /** The column number. */
+    uint32_t column;
+} pm_line_column_t;
+
+/**
+ * Initialize a new newline list with the given capacity. Returns true if the
+ * allocation of the offsets succeeds, otherwise returns false.
+ *
+ * @param list The list to initialize.
+ * @param start A pointer to the start of the source string.
+ * @param capacity The initial capacity of the list.
+ * @return True if the allocation of the offsets succeeds, otherwise false.
+ */
+bool pm_newline_list_init(pm_newline_list_t *list, const uint8_t *start, size_t capacity);
+
+/**
+ * Clear out the newlines that have been appended to the list.
+ *
+ * @param list The list to clear.
+ */
+void
+pm_newline_list_clear(pm_newline_list_t *list);
+
+/**
+ * Append a new offset to the newline list. Returns true if the reallocation of
+ * the offsets succeeds (if one was necessary), otherwise returns false.
+ *
+ * @param list The list to append to.
+ * @param cursor A pointer to the offset to append.
+ * @return True if the reallocation of the offsets succeeds (if one was
+ *     necessary), otherwise false.
+ */
+bool pm_newline_list_append(pm_newline_list_t *list, const uint8_t *cursor);
+
+/**
+ * Returns the line of the given offset. If the offset is not in the list, the
+ * line of the closest offset less than the given offset is returned.
+ *
+ * @param list The list to search.
+ * @param cursor A pointer to the offset to search for.
+ * @param start_line The line to start counting from.
+ * @return The line of the given offset.
+ */
+int32_t pm_newline_list_line(const pm_newline_list_t *list, const uint8_t *cursor, int32_t start_line);
+
+/**
+ * Returns the line and column of the given offset. If the offset is not in the
+ * list, the line and column of the closest offset less than the given offset
+ * are returned.
+ *
+ * @param list The list to search.
+ * @param cursor A pointer to the offset to search for.
+ * @param start_line The line to start counting from.
+ * @return The line and column of the given offset.
+ */
+pm_line_column_t pm_newline_list_line_column(const pm_newline_list_t *list, const uint8_t *cursor, int32_t start_line);
+
+/**
+ * Free the internal memory allocated for the newline list.
+ *
+ * @param list The list to free.
+ */
+void pm_newline_list_free(pm_newline_list_t *list);
+
+#endif

data/lib/datadog/ruby_core_source/ruby-4.0.0-preview2/prism/util/pm_string.h

@@ -0,0 +1,200 @@
+/**
+ * @file pm_string.h
+ *
+ * A generic string type that can have various ownership semantics.
+ */
+#ifndef PRISM_STRING_H
+#define PRISM_STRING_H
+
+#include "prism/defines.h"
+
+#include <assert.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <string.h>
+
+// The following headers are necessary to read files using demand paging.
+#ifdef _WIN32
+#include <windows.h>
+#elif defined(_POSIX_MAPPED_FILES)
+#include <fcntl.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#elif defined(PRISM_HAS_FILESYSTEM)
+#include <fcntl.h>
+#include <sys/stat.h>
+#endif
+
+/**
+ * A generic string type that can have various ownership semantics.
+ */
+typedef struct {
+    /** A pointer to the start of the string. */
+    const uint8_t *source;
+
+    /** The length of the string in bytes of memory. */
+    size_t length;
+
+    /** The type of the string. This field determines how the string should be freed. */
+    enum {
+        /** This string is a constant string, and should not be freed. */
+        PM_STRING_CONSTANT,
+
+        /** This is a slice of another string, and should not be freed. */
+        PM_STRING_SHARED,
+
+        /** This string owns its memory, and should be freed using `pm_string_free()`. */
+        PM_STRING_OWNED,
+
+#ifdef PRISM_HAS_MMAP
+        /** This string is a memory-mapped file, and should be freed using `pm_string_free()`. */
+        PM_STRING_MAPPED
+#endif
+    } type;
+} pm_string_t;
+
+/**
+ * Returns the size of the pm_string_t struct. This is necessary to allocate the
+ * correct amount of memory in the FFI backend.
+ *
+ * @return The size of the pm_string_t struct.
+ */
+PRISM_EXPORTED_FUNCTION size_t pm_string_sizeof(void);
+
+/**
+ * Defines an empty string. This is useful for initializing a string that will
+ * be filled in later.
+ */
+#define PM_STRING_EMPTY ((pm_string_t) { .type = PM_STRING_CONSTANT, .source = NULL, .length = 0 })
+
+/**
+ * Initialize a shared string that is based on initial input.
+ *
+ * @param string The string to initialize.
+ * @param start The start of the string.
+ * @param end The end of the string.
+ */
+void pm_string_shared_init(pm_string_t *string, const uint8_t *start, const uint8_t *end);
+
+/**
+ * Initialize an owned string that is responsible for freeing allocated memory.
+ *
+ * @param string The string to initialize.
+ * @param source The source of the string.
+ * @param length The length of the string.
+ */
+void pm_string_owned_init(pm_string_t *string, uint8_t *source, size_t length);
+
+/**
+ * Initialize a constant string that doesn't own its memory source.
+ *
+ * @param string The string to initialize.
+ * @param source The source of the string.
+ * @param length The length of the string.
+ */
+void pm_string_constant_init(pm_string_t *string, const char *source, size_t length);
+
+/**
+ * Represents the result of calling pm_string_mapped_init or
+ * pm_string_file_init. We need this additional information because there is
+ * not a platform-agnostic way to indicate that the file that was attempted to
+ * be opened was a directory.
+ */
+typedef enum {
+    /** Indicates that the string was successfully initialized. */
+    PM_STRING_INIT_SUCCESS = 0,
+    /**
+     * Indicates a generic error from a string_*_init function, where the type
+     * of error should be read from `errno` or `GetLastError()`.
+     */
+    PM_STRING_INIT_ERROR_GENERIC = 1,
+    /**
+     * Indicates that the file that was attempted to be opened was a directory.
+     */
+    PM_STRING_INIT_ERROR_DIRECTORY = 2
+} pm_string_init_result_t;
+
+/**
+ * Read the file indicated by the filepath parameter into source and load its
+ * contents and size into the given `pm_string_t`. The given `pm_string_t`
+ * should be freed using `pm_string_free` when it is no longer used.
+ *
+ * We want to use demand paging as much as possible in order to avoid having to
+ * read the entire file into memory (which could be detrimental to performance
+ * for large files). This means that if we're on windows we'll use
+ * `MapViewOfFile`, on POSIX systems that have access to `mmap` we'll use
+ * `mmap`, and on other POSIX systems we'll use `read`.
+ *
+ * @param string The string to initialize.
+ * @param filepath The filepath to read.
+ * @return The success of the read, indicated by the value of the enum.
+ *
+ * \public \memberof pm_string_t
+ */
+PRISM_EXPORTED_FUNCTION pm_string_init_result_t pm_string_mapped_init(pm_string_t *string, const char *filepath);
+
+/**
+ * Read the file indicated by the filepath parameter into source and load its
+ * contents and size into the given `pm_string_t`. The given `pm_string_t`
+ * should be freed using `pm_string_free` when it is no longer used.
+ *
+ * @param string The string to initialize.
+ * @param filepath The filepath to read.
+ * @return The success of the read, indicated by the value of the enum.
+ *
+ * \public \memberof pm_string_t
+ */
+PRISM_EXPORTED_FUNCTION pm_string_init_result_t pm_string_file_init(pm_string_t *string, const char *filepath);
+
+/**
+ * Ensure the string is owned. If it is not, then reinitialize it as owned and
+ * copy over the previous source.
+ *
+ * @param string The string to ensure is owned.
+ */
+void pm_string_ensure_owned(pm_string_t *string);
+
+/**
+ * Compare the underlying lengths and bytes of two strings. Returns 0 if the
+ * strings are equal, a negative number if the left string is less than the
+ * right string, and a positive number if the left string is greater than the
+ * right string.
+ *
+ * @param left The left string to compare.
+ * @param right The right string to compare.
+ * @return The comparison result.
+ */
+int pm_string_compare(const pm_string_t *left, const pm_string_t *right);
+
+/**
+ * Returns the length associated with the string.
+ *
+ * @param string The string to get the length of.
+ * @return The length of the string.
+ *
+ * \public \memberof pm_string_t
+ */
+PRISM_EXPORTED_FUNCTION size_t pm_string_length(const pm_string_t *string);
+
+/**
+ * Returns the start pointer associated with the string.
+ *
+ * @param string The string to get the start pointer of.
+ * @return The start pointer of the string.
+ *
+ * \public \memberof pm_string_t
+ */
+PRISM_EXPORTED_FUNCTION const uint8_t * pm_string_source(const pm_string_t *string);
+
+/**
+ * Free the associated memory of the given string.
+ *
+ * @param string The string to free.
+ *
+ * \public \memberof pm_string_t
+ */
+PRISM_EXPORTED_FUNCTION void pm_string_free(pm_string_t *string);
+
+#endif

data/lib/datadog/ruby_core_source/ruby-4.0.0-preview2/prism/util/pm_strncasecmp.h

@@ -0,0 +1,32 @@
+/**
+ * @file pm_strncasecmp.h
+ *
+ * A custom strncasecmp implementation.
+ */
+#ifndef PRISM_STRNCASECMP_H
+#define PRISM_STRNCASECMP_H
+
+#include "prism/defines.h"
+
+#include <ctype.h>
+#include <stddef.h>
+#include <stdint.h>
+
+/**
+ * Compare two strings, ignoring case, up to the given length. Returns 0 if the
+ * strings are equal, a negative number if string1 is less than string2, or a
+ * positive number if string1 is greater than string2.
+ *
+ * Note that this is effectively our own implementation of strncasecmp, but it's
+ * not available on all of the platforms we want to support so we're rolling it
+ * here.
+ *
+ * @param string1 The first string to compare.
+ * @param string2 The second string to compare
+ * @param length The maximum number of characters to compare.
+ * @return 0 if the strings are equal, a negative number if string1 is less than
+ *     string2, or a positive number if string1 is greater than string2.
+ */
+int pm_strncasecmp(const uint8_t *string1, const uint8_t *string2, size_t length);
+
+#endif

data/lib/datadog/ruby_core_source/ruby-4.0.0-preview2/prism/util/pm_strpbrk.h

@@ -0,0 +1,46 @@
+/**
+ * @file pm_strpbrk.h
+ *
+ * A custom strpbrk implementation.
+ */
+#ifndef PRISM_STRPBRK_H
+#define PRISM_STRPBRK_H
+
+#include "prism/defines.h"
+#include "prism/diagnostic.h"
+#include "prism/parser.h"
+
+#include <stddef.h>
+#include <string.h>
+
+/**
+ * Here we have rolled our own version of strpbrk. The standard library strpbrk
+ * has undefined behavior when the source string is not null-terminated. We want
+ * to support strings that are not null-terminated because pm_parse does not
+ * have the contract that the string is null-terminated. (This is desirable
+ * because it means the extension can call pm_parse with the result of a call to
+ * mmap).
+ *
+ * The standard library strpbrk also does not support passing a maximum length
+ * to search. We want to support this for the reason mentioned above, but we
+ * also don't want it to stop on null bytes. Ruby actually allows null bytes
+ * within strings, comments, regular expressions, etc. So we need to be able to
+ * skip past them.
+ *
+ * Finally, we want to support encodings wherein the charset could contain
+ * characters that are trailing bytes of multi-byte characters. For example, in
+ * Shift-JIS, the backslash character can be a trailing byte. In that case we
+ * need to take a slower path and iterate one multi-byte character at a time.
+ *
+ * @param parser The parser.
+ * @param source The source to search.
+ * @param charset The charset to search for.
+ * @param length The maximum number of bytes to search.
+ * @param validate Whether to validate that the source string is valid in the
+ *     current encoding of the parser.
+ * @return A pointer to the first character in the source string that is in the
+ *     charset, or NULL if no such character exists.
+ */
+const uint8_t * pm_strpbrk(pm_parser_t *parser, const uint8_t *source, const uint8_t *charset, ptrdiff_t length, bool validate);
+
+#endif

data/lib/datadog/ruby_core_source/ruby-4.0.0-preview2/prism/version.h

@@ -0,0 +1,29 @@
+/**
+ * @file version.h
+ *
+ * The version of the Prism library.
+ */
+#ifndef PRISM_VERSION_H
+#define PRISM_VERSION_H
+
+/**
+ * The major version of the Prism library as an int.
+ */
+#define PRISM_VERSION_MAJOR 1
+
+/**
+ * The minor version of the Prism library as an int.
+ */
+#define PRISM_VERSION_MINOR 6
+
+/**
+ * The patch version of the Prism library as an int.
+ */
+#define PRISM_VERSION_PATCH 0
+
+/**
+ * The version of the Prism library as a constant string.
+ */
+#define PRISM_VERSION "1.6.0"
+
+#endif

data/lib/datadog/ruby_core_source/ruby-4.0.0-preview2/prism_compile.h

@@ -0,0 +1,106 @@
+#include "prism/prism.h"
+#include "ruby/encoding.h"
+
+/**
+ * the getlocal and setlocal instructions require two parameters. level is how
+ * many hops up the iseq stack one needs to go before finding the correct local
+ * table. The index is the index in that table where our variable is.
+ *
+ * Because these are always calculated and used together, we'll bind them
+ * together as a tuple.
+ */
+typedef struct pm_local_index_struct {
+    int index, level;
+} pm_local_index_t;
+
+// A declaration for the struct that lives in compile.c.
+struct iseq_link_anchor;
+
+// ScopeNodes are helper nodes, and will never be part of the AST. We manually
+// declare them here to avoid generating them.
+typedef struct pm_scope_node {
+    pm_node_t base;
+    struct pm_scope_node *previous;
+    pm_node_t *ast_node;
+    pm_node_t *parameters;
+    pm_node_t *body;
+    pm_constant_id_list_t locals;
+
+    const pm_parser_t *parser;
+    rb_encoding *encoding;
+
+    /**
+     * This is a pointer to the list of script lines for the ISEQs that will be
+     * associated with this scope node. It is only set if
+     * RubyVM.keep_script_lines is true. If it is set, it will be set to a
+     * pointer to an array that is always stack allocated (so no GC marking is
+     * needed by this struct). If it is not set, it will be NULL. It is
+     * inherited by all child scopes.
+     */
+    VALUE *script_lines;
+
+    /**
+     * This is the encoding of the actual filepath object that will be used when
+     * a __FILE__ node is compiled or when the path has to be set on a syntax
+     * error.
+     */
+    rb_encoding *filepath_encoding;
+
+    // The size of the local table on the iseq which includes locals and hidden
+    // variables.
+    int local_table_for_iseq_size;
+
+    ID *constants;
+    st_table *index_lookup_table;
+
+    // The current coverage setting, passed down through the various scopes.
+    int coverage_enabled;
+
+    /**
+     * This will only be set on the top-level scope node. It will contain all of
+     * the instructions pertaining to BEGIN{} nodes.
+     */
+    struct iseq_link_anchor *pre_execution_anchor;
+} pm_scope_node_t;
+
+void pm_scope_node_init(const pm_node_t *node, pm_scope_node_t *scope, pm_scope_node_t *previous);
+void pm_scope_node_destroy(pm_scope_node_t *scope_node);
+
+typedef struct {
+    /** The parser that will do the actual parsing. */
+    pm_parser_t parser;
+
+    /** The options that will be passed to the parser. */
+    pm_options_t options;
+
+    /** The input that represents the source to be parsed. */
+    pm_string_t input;
+
+    /** The resulting scope node that will hold the generated AST. */
+    pm_scope_node_t node;
+
+    /** Whether or not this parse result has performed its parsing yet. */
+    bool parsed;
+} pm_parse_result_t;
+
+#define PM_SPECIAL_CONSTANT_FLAG ((pm_constant_id_t)(1 << 31))
+#define PM_CONSTANT_AND ((pm_constant_id_t)(idAnd | PM_SPECIAL_CONSTANT_FLAG))
+#define PM_CONSTANT_DOT3 ((pm_constant_id_t)(idDot3 | PM_SPECIAL_CONSTANT_FLAG))
+#define PM_CONSTANT_MULT ((pm_constant_id_t)(idMULT | PM_SPECIAL_CONSTANT_FLAG))
+#define PM_CONSTANT_POW ((pm_constant_id_t)(idPow | PM_SPECIAL_CONSTANT_FLAG))
+
+VALUE pm_load_file(pm_parse_result_t *result, VALUE filepath, bool load_error);
+VALUE pm_parse_file(pm_parse_result_t *result, VALUE filepath, VALUE *script_lines);
+VALUE pm_load_parse_file(pm_parse_result_t *result, VALUE filepath, VALUE *script_lines);
+VALUE pm_parse_string(pm_parse_result_t *result, VALUE source, VALUE filepath, VALUE *script_lines);
+VALUE pm_parse_stdin(pm_parse_result_t *result);
+void pm_options_version_for_current_ruby_set(pm_options_t *options);
+void pm_parse_result_free(pm_parse_result_t *result);
+
+rb_iseq_t *pm_iseq_new(pm_scope_node_t *node, VALUE name, VALUE path, VALUE realpath, const rb_iseq_t *parent, enum rb_iseq_type, int *error_state);
+rb_iseq_t *pm_iseq_new_top(pm_scope_node_t *node, VALUE name, VALUE path, VALUE realpath, const rb_iseq_t *parent, int *error_state);
+rb_iseq_t *pm_iseq_new_main(pm_scope_node_t *node, VALUE path, VALUE realpath, const rb_iseq_t *parent, int opt, int *error_state);
+rb_iseq_t *pm_iseq_new_eval(pm_scope_node_t *node, VALUE name, VALUE path, VALUE realpath, int first_lineno, const rb_iseq_t *parent, int isolated_depth, int *error_state);
+rb_iseq_t *pm_iseq_new_with_opt(pm_scope_node_t *node, VALUE name, VALUE path, VALUE realpath, int first_lineno, const rb_iseq_t *parent, int isolated_depth, enum rb_iseq_type, const rb_compile_option_t *option, int *error_state);
+
+VALUE pm_iseq_compile_node(rb_iseq_t *iseq, pm_scope_node_t *node);